Yesterday's numbers, on your phone by eight
A daily page the whole team opens before the first meeting, refreshed without anyone exporting anything.
Data · AI · Technology. Turning Data into Direction.
Learn data by solving real problems
Pick the tool or language you want. Every lesson starts from a problem people face at work, not from a list of features. Short episodes, taught by an architect who has built these for global businesses for over a decade.
- 6 episodes ready today
- Real use cases
- 7 learning paths
Move across the field, click to send a wave
What you will be able to do
Real problems you will solve
These are everyday jobs in any business. Each one is a lesson, and each uses the tools shown.
Two systems disagree. Find out why in minutes.
Match the till against the bank, or bookings against invoices, and list exactly which lines do not agree.
Forty spreadsheets become one table
Every branch sends a file in its own layout. A short script reads them all and stacks them cleanly, every month.
Ask your data a question in plain words
Type "which products lost margin this quarter?" and get an answer you can check against the source.
A nightly pipeline that tells you when it breaks
Data lands, gets cleaned and is ready by morning. If a step fails, you hear about it before the report does.
Each manager sees only their own numbers
One report for the whole company, with every region, store or department limited to its own slice.
An assistant that answers from your own documents
Policies, contracts and manuals become something staff can question in plain words, with the source shown.
An agent that prepares the weekly report
It gathers the figures, checks them against last week, drafts the commentary and waits for your approval.
A data and AI plan the board will fund
One page that says where the business is, what to build first and how the value will be measured.
Your choice
Choose what you want to learn
Pick a tool or a language. Each is taught through the problems it solves at work.
Drag to turn it. Tap a point to choose.
Beginner, ready today
Data and BI from zero
Understand what business intelligence is, and read any report with confidence.
People use it to
- Read any dashboard and spot what it is not telling you
- Brief an analyst so you get the right report the first time
- Sketch a page your team will open every morning
Beginner, next course
SQL for beginners
Ask a database your own questions instead of waiting for an export.
People use it to
- Find the customers who have not ordered in ninety days
- Check yesterday's sales in two systems and list what differs
- Build a month-end total from raw transactions, and prove it
Intermediate, on the roadmap
Power BI, from first report to a model people trust
Build a report that answers a real question, on a model that gives the same number every time.
People use it to
- A daily sales page the whole team opens before nine
- Budget against actual by department, with the detail one click away
- One report where each manager sees only their own stores
Beginner, on the roadmap
Python for analysts
Replace the manual parts of your week with scripts you understand.
People use it to
- Stack forty Excel files into one clean table in seconds
- Get an email the moment a number crosses a limit
- Clean the same messy export automatically, every day
Intermediate, on the roadmap
Microsoft Fabric, the working parts
Know what each part of Fabric is for, and move one dataset through all of them.
People use it to
- One place where sales, finance and operations data meet
- Reports that stay current without overnight copy jobs
- Move a dataset from raw file to finished report, end to end
Intermediate, on the roadmap
AWS for data teams
Find your way around the handful of AWS services a data team actually uses.
People use it to
- Land files from any system safely in one place
- Run a pipeline every night and hear about it if it fails
- Query years of history without managing a server
Intermediate, on the roadmap
Azure for data apps
Run a data app on Azure without guessing at identity, secrets or cost.
People use it to
- Let an app read data with no password in the code
- Host a small internal data tool securely
- Know what the cloud bill will be before it arrives
Advanced, on the roadmap
Fabric architecture, designing the platform
Lay out a Fabric estate that stays understandable as it grows.
People use it to
- Decide who sees what before the first report is built
- Release changes without breaking what people rely on
- A platform three teams can share without collisions
Intermediate, on the roadmap
AI for data and business teams
Put AI to work on real business questions, and know when not to trust the answer.
People use it to
- Ask a question in plain words and get a number you can check
- Turn a fifty-page pack into a paragraph you can trust
- Draft the monthly commentary from the figures themselves
Intermediate, on the roadmap
Large language models, explained for builders and leaders
Understand how language models work well enough to choose one, ground it and judge its answers.
People use it to
- Explain to a board what a language model can and cannot do
- Build an assistant that answers from your own documents
- Compare models on your own questions before paying for one
Advanced, on the roadmap
Agentic AI, from assistant to colleague
Design AI agents that plan, use tools and finish real tasks, with people in control.
People use it to
- An agent that gathers the numbers and drafts the weekly report
- Sort incoming requests and route each one to the right team
- Decide exactly where a person must approve before an agent acts
Advanced, on the roadmap
Leading with data and AI
Set direction for a data and AI team: what to build, how to govern it and how to prove its value.
People use it to
- Write a one-page data and AI strategy the board will fund
- Decide what to build, what to buy and what to stop
- Set up a team where every number has an owner
Start here
One complete course, ready today
6 short episodes for complete beginners. No software to install.
BI basics, beginner
Data and BI from zero
Understand what business intelligence is, and read any report with confidence.
- What BI actually isReadyFrom raw records to a decision, in one picture.
- Rows, columns and tablesReadyHow data is stored, and why a tidy table beats a pretty one.
- Measures and dimensionsReadyThe numbers you add up, and the labels you slice them by.
- Reading a chart honestlyReadyAxes, scales and the tricks a chart can play on you.
- Where the data comes fromReadySource systems, exports, refreshes and what "live" really means.
- Your first dashboard, on paperReadySketch a page that answers one question before touching a tool.
How it fits together
One stack, from raw data to a trusted answer
This is the whole map. Every course on the roadmap teaches one layer of it, from the bottom up.
Drag to turn it. Point at a layer to light it up.
- 4
Answers
Ask a question in plain words and trust what comes back.
- 3
Model and report
One definition of every number, on a page people open every morning.
- 2
Platform
Where the data lands, gets cleaned and is kept in one copy.
- 1
Cloud and code
The services and scripts that move data without anyone doing it by hand.
Learning paths
Not sure where to start? Follow a path
Each path is a roadmap: courses in the right order for one goal. The filled step is ready today.
Beginner path
Start from zero
New to data, comfortable in a spreadsheet.
You finish able to: Read a database, build a first report and explain what it shows.
Intermediate path
Analyst to BI developer
Analysts who build reports and want to own the model behind them.
You finish able to: A governed Power BI model, fed by scripts instead of by hand.
Intermediate path
Data platform builder
People who will run the pipes behind the reports.
You finish able to: Land, clean and serve data on a cloud platform.
Every second Thursday
The Helmcraft Letter
One idea on data, AI and leading with numbers. Four minutes, one idea, no filler.
14 Nov 2026Why HelmcraftWhat this is, who it is for and what to expect from it.12 Nov 2026A gap is not timing until you prove itThe most expensive word in a reconciliation, and three habits that retire it.29 Oct 2026One number, one ownerWhy every figure that leadership reads needs a single place it comes from.
Being written
- What an assistant must know before it answers
- Close on day five, not day fifteen
- The cost of a number nobody questions
- Teaching a team to read a variance
Who writes this
Taught from the work, not the manual
I am Shri. For over a decade I have designed data platforms and analytics for global businesses in retail, manufacturing, healthcare, consumer goods and hospitality. Today I lead data analytics and AI for an international group. This is where I teach what that work has taught me, in order.
- Microsoft Fabric
- Power BI
- Azure
- AWS
- Databricks
- Snowflake
- SQL
- Python
- Applied AI
- Based in
- Dubai
- Experience
- Since 2014, seven industries
- Publishes
- Every second Thursday